Independent evals, research and analysis for voice AI. We evaluate models in the conditions they get deployed in, and work with labs to fix what breaks.
San Francisco
Joined May 2026
- Tweets264
- Following1
- Followers1.4K
- Likes107
Voice Arena retweeted
Maya-2 voice model weights are officially live! 🇮🇳🇮🇳
The frontier voice model for India is now completely open-weight.
• Outperforms ElevenLabs v3, Sarvam Bulbul v3, and Cartesia Sonic 3.5
• 1.5B+ people covered across all 28 states and 8 union territories
• SOTA voice synthesis across 11 official languages
• 20 native speakers pre-loaded, ready out of the box
• Best-in-class zero-shot voice cloning
Built by a team of five from the land of languages.
Go build and have fun!
Voice Arena retweeted
Nemotron 3 Diarization ranked #1 of 12 systems in @voicearena_ai's initial Diarization-Bench results. Its 14.72% error rate was ~24% lower than the runner-up.
Here’s a four-speaker comparison from the benchmark.
Full results: voicearena.com/diarization-b…
Voice Arena retweeted
Excited to share some great news. @voicearena_ai is the Diamond Sponsor of Interspeech 2026.
I’ve been attending Interspeech for the last 2 years. My journey of falling in love with research and solving bottlenecks in voice ai started here.
We’ll be sharing our latest evaluations, datasets and research with the community.
If you are going to be there, do come by!
📍 Booth D1, ICC Sydney
📅 28 September–1 October
Voice Arena retweeted
This is my favourite graph from @voicearena_ai's Jarvis Bench. If you've wondered why voice agents aren't being used more, this explain it.
The best models are getting the job done. But they still don't sound human.
Voice Arena retweeted
Google's Gemini 3.8 Live debuts at #2 on Voice Arena's Jarvis Bench, the leading model across the leaderboard at 991 naturalness Elo.
Multi-turn real-time calls, human callers, human raters, and a human agent on the board. voicearena.com
@GoogleDeepMind
Voice Arena retweeted
Our mission is to ensure that machines talk like humans and we’re excited to be building standards that push that forward @voicearena_ai nitter.cf/i/trending/21001367929…
Voice Arena retweeted
Introducing Jarvis Bench v0.5, @voicearena_ai's conversational agent benchmark.
We've been obsessed with one question at VoiceArena: why do voice agent demos sound incredible, benchmarks say models are near-perfect, and yet you probably didn't have a single real conversation with a voice agent in the last 24 hours?
Here's what's different about the Jarvis Bench. Real humans have live conversations with voice agents. A second group of humans blind-votes pairwise on two questions: which sounded more human, which got the job done. And one of the "agents" on the leaderboard is a human.
Voice Arena retweeted
Hosted the first edition of Voice Research Club in SF yesterday with @rjrshr and @nishnik!
We’ll be hosting this monthly. The next edition is on October 8.
VRC is a volunteer-driven community. If you’d like to help shape future editions - from deciding which papers get presented to helping organize the sessions - DM any of us.
@voicearena_ai
Voice Arena retweeted
Cartesia Sonic-3.6 is the #1 TTS model on the Voice Arena US English 🇺🇸 Leaderboard.
It debuts at 1085 Elo, in a statistical tie with Google DeepMind's Gemini 3.1 Flash TTS, and ahead of its own Sonic-3.5, Simba 3.2 and Grok TTS. It holds the top spot as a real-time model.
Sonic-3.6 is the latest TTS model from @cartesia. The model has been highly preferred among raters on @voicearena_ai.
Results backed by 1,400+ blind, head-to-head listener votes across two languages 🧵
Voice Arena retweeted
🚨 The first domain-specific TTS leaderboards are live.
Today we're releasing Banking, Insurance, and Home Services leaderboards on VoiceArena US English - the first standing TTS benchmarks anywhere that rank models on the exact phrasing they'll be deployed to handle in production.
Every other public TTS leaderboard pools all use cases into one score. A model that wins overall can still be the worst on a fraud alert or a claim filing.
Three leaderboards live. Many more coming.
🧵👇
Voice Arena retweeted
1/Arabic just got its scoreboard. 🎙️
We ran today's leading TTS models through blind A/B trials - native Arabic listeners, two voices at a time, no labels.
Here's where every model landed 👇
Voice Arena retweeted
We will be extending the TTS eval to 8 additional domains including Banking, Telecom, Healthcare, Software % SaaS, Retail, Insurance, Airlines and Restaurants. @voicearena_ai
